Loading ...
Sorry, an error occurred while loading the content.
 

Re: Listening to Anderson on

Expand Messages
  • David J Anderson
    A mature organization has a failure tolerant culture. It innovates in its products and its processes. It is continuously improving. And it is doing all this in
    Message 1 of 8 , Nov 17, 2008
      A mature organization has a failure tolerant culture. It innovates in
      its products and its processes. It is continuously improving. And it
      is doing all this in an objective and entirely defensible fashion.
      Because the innovation opportunities are based on objective evaluation
      of data, there is an institutional memory developed about what worked
      or didn't and why it may have worked.

      Mature organizations use leading indicators to make predictive
      management interventions that mean they react to situations faster.
      Often improving quality or lead time as a result.

      Mature organizations have processes with low degrees of variation.

      My observation of most agile teams is that they do not innovate in the
      broad sense. They tend to optimize around a process definition, like
      XP or Scrum. They are not failure tolerant. And they tend to be mainly
      subjective in their decision making and heavily reliant on tactic
      knowledge which impedes the development of an institutional memory.

      Agile organizations tend to be entirely reactive. In fact, some of the
      more extreme folks in the community believe that we should only ever
      be reactive and that being reactive rather than proactive or
      predictive is a core essence of "agile."

      Immature organizations have processes with high degrees of
      variability. In this area, agile teams vary greatly. Some of them are
      very good at reducing variability - using a story template and
      promiscuous pair programming for example are two very good ways to
      reduce variability.

      An organization that uses leading indicators properly and has a
      "predictable" process (that is one that exhibits low degrees of
      variation), will outperform a purely reactive organization.

      So while I have heard of agile teams that appear to exhibit high
      maturity behaviors - objectivity, use of leading indicators, low
      degrees of variability, and (maybe, just maybe) continuous improvement
      in a failure tolerant culture, I have not heard of one that existed
      without the direct leadership of one of the personalities in our
      community. At this point, it is impossible to take the "David" or
      "Jeff" or "Israel" factor out of the achievement of the high maturity.

      David
      http://www.agilemanagement.net/


      --- In kanbandev@yahoogroups.com, "Matt Heusser" <matt.heusser@...> wrote:
      >
      > I'm listening to the interview of David Anderson for the Agile Dev
      Practices
      > Conference:
      >
      >
      http://www.stickyminds.com/ControlImages/Stickyminds/Image/Podcast/Audio/ADP08Anderson.mp3
      >
      > Toward the end, David says that success at agile in the large requires
      > culture change and Organizationals Process Maturity.
      >
      > David, I've read a lot (lot) of material on KanBan in the past
      weeks. I've
      > watched the videos, reads the blogs, and read the whitepapers.
      >
      > Can you tell us a little bit about what you mean by "Organizational
      > Maturity" - beyond the CMMI sense. What does a mature organization look
      > like? How does it behave? How is it different than a non-mature
      company
      > that is "just" "doing agile"?
      >
      > If you mean in the CMMI sense of mature, could you elaborate on the same
      > questions for us for our benefit?
      >
      > regards,
      >
      > --
      > Matthew Heusser,
      > Blog: http://xndev.blogspot.com
      > Twitter: http://www.twitter.com/mheusser
      > Skype: mattheusser
      >
    • Brian Marick
      ... That s odd. My observation is that good Agile teams are relatively blasé about failure because they re confident that they can recover from it. (That is,
      Message 2 of 8 , Nov 18, 2008
        On Nov 18, 2008, at 1:45 AM, David J Anderson wrote:
        >
        >
        > My observation of most agile teams is that they [...]
        > are not failure tolerant.

        That's odd. My observation is that good Agile teams are relatively
        blasé about failure because they're confident that they can recover
        from it. (That is, a failure won't lead to a huge variation in
        productivity etc., so they can afford to experiment.) Can you say more
        about that?

        > And they tend to be mainly
        > subjective in their decision making and heavily reliant on tactic
        > knowledge which impedes the development of an institutional memory.


        This I agree on. (I'm assuming "tactic" is a typo for "tacit".) But a
        great many craft-like professions rely heavily on tacit knowledge for
        their institutional memory. Examples:

        1. You don't learn how to do proofs by reading proofs, because a proof
        contains no history of its development. The craft of "mathematician"
        isn't (much) taught so much as absorbed in apprenticeship (barring the
        outlier genius like Ramanujan).

        2. Physicians have a solid institutional memory (in both the senses of
        "firmly established" and "hard to move"). It is largely tacit and
        greatly subjective. That's by no means an unalloyed good - my wife is
        a big fan of what they call "evidence-based medicine" - but it's still
        a successful practical field.

        My point is that saying immature is to mature as subjective is to
        objective or as tacit is to explicit condemns some pretty effective
        professions to immaturity. And one could snarkily point out that bond
        rating agencies, investment banks, and quantitative hedge funds did
        quite well according to this definition:

        > A mature organization has a failure tolerant culture. It innovates in
        > its products and its processes. It is continuously improving. And it
        > is doing all this in an objective and entirely defensible fashion.
        > Because the innovation opportunities are based on objective evaluation
        > of data, there is an institutional memory developed about what worked
        > or didn't and why it may have worked.


        Their models were objective and entirely defensible given the
        historical data. They just had problems when real distributions turned
        out to be long-tailed or housing prices - which had not declined for
        the memory-period - declined.

        A larger point is that "maturity" is a word that fights against
        reasoned argument because it has too many connotations. It invites
        subjective reactions, not objective. If, by "maturity", you mean "data-
        driven, variation-reducing, and change-seeking", I think you should
        say that. Then people will ask "What do you mean by that?" instead of
        automatically equating what you're talking about with "good", "adult",
        "responsible", and other qualitatively warm-n-fuzzy adjectives.

        -----
        Brian Marick, independent consultant
        Mostly on agile methods with a testing slant
        www.exampler.com, www.exampler.com/blog, www.twitter.com/marick
      • David J Anderson
        Brian, Your edit below removed the context that qualified the sentence. The point is that I see teams who focus on getting very good at scrum or on pursuing
        Message 3 of 8 , Nov 18, 2008
          Brian,

          Your edit below removed the context that qualified the sentence. The
          point is that I see teams who focus on getting "very good at scrum" or
          on pursuing "pure XP" (remember what Dan North had to say about this
          in his key note at Agile Practices?). They do not innovate out of the
          bounds of their chosen methodology.

          In that context they are not tolerant of innovation. Perhaps not
          tolerant of failure is the wrong characterization? Perhaps it's a
          belligerence or dogma that is the problem? Or narrow-mindedness?

          I agreed with much of the rest of your post, though I do not believe
          the medical profession would resonate with the idea that they are
          subjective and have a tacit knowledge-based institutional memory.
          There is lots of science behind medical recommendations and lots of
          data gathered. Ask a cancer patient or an oncologist for example?

          David
          http://www.agilemanagement.net/



          --- In kanbandev@yahoogroups.com, Brian Marick <marick@...> wrote:
          >
          >
          > On Nov 18, 2008, at 1:45 AM, David J Anderson wrote:
          > >
          > >
          > > My observation of most agile teams is that they [...]
          > > are not failure tolerant.
          >
          > That's odd. My observation is that good Agile teams are relatively
          > blasé about failure because they're confident that they can recover
          > from it. (That is, a failure won't lead to a huge variation in
          > productivity etc., so they can afford to experiment.) Can you say more
          > about that?
          >
        • Karl Scotland
          Nice. You should blog this! One comment below. 2008/11/18 David J Anderson ... I think this observation ties into the various blogs
          Message 4 of 8 , Nov 18, 2008
            Nice.  You should blog this!  One comment below.

            2008/11/18 David J Anderson <netherby_uk@...>

            My observation of most agile teams is that they do not innovate in the
            broad sense. They tend to optimize around a process definition, like
            XP or Scrum. 

            I think this observation ties into the various blogs etc recently about Scrum/Agile failing. (http://www.infoq.com/news/2008/11/decline-of-agile).  Scum and other Agile methods were innovations when they emerged, but are now seen as process definitions.

            --
            Karl Scotland
            Agile Coach
            http://availagility.wordpress.com/
          • Christophe Louvion
            My experience resonates with D s comment here. I have increased team productivity (value delivery) many folds... to see this success vanish when I left a
            Message 5 of 8 , Nov 19, 2008
              My experience resonates with D's comment here.
              I have increased team productivity (value delivery) many folds... to see
              this success vanish when I left a company.
              While this guarantees good paychecks, experienced leaders move out of
              companies for better opportunities, retirement, illness etc.

              Most agile implementations focus on the practices which lead to
              fantastic short term benefits but don't hold over time.
              I guess focusing on organization maturity is about creating an
              enterprise wide (and deep) culture that does not count on a few leaders
              to keep progress going, but teaches continuous improvement all around.
              It is a slower, harder yet sustainable approach.

              In my new organization, most teams started with scrum and some XP
              colors, some switched to kanban, some not.
              While we still have a lot of work to do, we are not taking anything for
              granted, and allow ourselves to experiment around anything (reporting
              structure, iterations length / no iteration, rotating roles, backlog
              size, you name it).
              Many teams fake doing agile by copycatting scrum, not doing daily stand
              ups, retrospective, and running mini waterfall so someone can tell their
              boss they are now agile.
              I think we're doing doing something different. Trying to really create a
              change structure at all levels. My departure from my company will be the
              ultimate test to the success I guess.

              Any suggestion to evaluate if the culture actually is what we try to
              create, before I actually leave?


              >I have not heard of one that existed
              >without the direct leadership of one of the personalities in our
              >community. At this point, it is impossible to take the "David" or
              >"Jeff" or "Israel" factor out of the achievement of the high maturity.
            • Brian Marick
              ... It s a little hackneyed, but you could describe XP or Scrum as having moved from Kuhn s revolutionary science to his normal science. In the latter
              Message 6 of 8 , Nov 19, 2008
                On Nov 18, 2008, at 9:55 AM, David J Anderson wrote:
                >
                > The
                > point is that I see teams who focus on getting "very good at scrum" or
                > on pursuing "pure XP" (remember what Dan North had to say about this
                > in his key note at Agile Practices?). They do not innovate out of the
                > bounds of their chosen methodology.

                It's a little hackneyed, but you could describe XP or Scrum as having
                moved from Kuhn's "revolutionary" science to his "normal" science. In
                the latter phase, you're applying known techniques and established
                patterns of thought to new-but-not-radically-new problems. That's a
                less judgmental, emotion-laden way of putting it than (as below )
                "intolerant of innovation" or referring to "dogma" or "narrow-
                mindedness". It's part of a process rather than a fault of particular
                bad people. Normal science is less flashy than Einstein or Newton, but
                it's needed.

                > In that context they are not tolerant of innovation. Perhaps not
                > tolerant of failure is the wrong characterization? Perhaps it's a
                > belligerence or dogma that is the problem? Or narrow-mindedness?

                I'm not actually a huge follower of Kuhn. When it comes to (quite
                possibly bogus) parallels between scientific methodology and software
                methodology, I make more use of Imre Lakatos' _Methodology of
                Scientific Research Programmes_. I wrote a long blog post about that
                before, so I won't repeat it here.
                <http://www.exampler.com/old-blog/2003/04/28/#lakatos>

                The reason I think it's relevant is because I don't think the idea of
                a methodology (or its followers) innovating outside its own bounds
                really stands up. I like Lakatos' idea of a methodology as a concerted
                extension of core postulates, and his notion of judging a research
                programme (or development of a software methodology) by how it reacts
                to counterexamples and contradictions. It's a shame he died while with
                his ideas still underexplained.

                > I agreed with much of the rest of your post, though I do not believe
                > the medical profession would resonate with the idea that they are
                > subjective and have a tacit knowledge-based institutional memory.
                > There is lots of science behind medical recommendations and lots of
                > data gathered. Ask a cancer patient or an oncologist for example?

                There was a fairly recent study about geographical patterns in
                treatments. (No reference, sorry.) To simplify: doctors in, say,
                Boston prescribe treatment X for ailment A far more often than
                treatment Y. But doctors in Atlanta prescribe treatment Y more often
                than X. Why? It's not the scientific evidence, equally available to
                both groups. If I recall correctly, the study concluded it was just
                locality - Boston doctors liked X, so new doctors in Boston came to
                treat the way doctors in Boston treat. That's institutional memory
                that's not knowledge-based.

                Sure, there's lots of data in medicine. For some field research on
                antibiotic effectiveness and cost-effectiveness (in something as
                relatively unimportant as cattle), my wife gathered 900,000 data
                points. Interestingly, her results confirmed common veterinarian
                practice, which was (and is) exactly the opposite of what smaller-
                scale laboratory trials indicated was proper. It's not scandalous that
                veterinarians ignored studies that contradicted their empirical
                observations, any more than it is that Newton ignored the Astronomer
                Royal's data that contradicted his theory of gravitation or that
                Eddington used judgment in correcting for errors in the famous
                observations of the 1919 total solar eclipse that confirmed general
                relativity. (More description and citations in link above).

                A great many physicians pride their field on being science-based - and
                it's certainly more science-based than ours! - but they still make off-
                label prescriptions.

                -----
                Brian Marick, independent consultant
                Mostly on agile methods with a testing slant
                www.exampler.com, www.exampler.com/blog, www.twitter.com/marick
              • David J Anderson
                The test I was taught at Sprint is the question, Have you worked yourself out of a job? i.e. are your direct reports doing your job for you and nothing has
                Message 7 of 8 , Nov 19, 2008
                  The test I was taught at Sprint is the question, "Have you worked
                  yourself out of a job?" i.e. are your direct reports doing your job
                  for you and nothing has changed?

                  However, this doesn't completely account for the acquiescence factor.
                  Are they doing it your way to please you?

                  And do you have any dissenters who will undermine your culture and
                  changes as soon as you leave?

                  These latter two are harder to answer. Both of them caught me out at
                  Corbis. And as a result the changes were not sustained despite the
                  high level of maturity that had been achieved at the point when I left.

                  Meanwhile, the workforce are left to lament the "good old days" when
                  they got a flawless release out every 2nd Wednesday.

                  David


                  --- In kanbandev@yahoogroups.com, Christophe Louvion
                  <chrislouvion@...> wrote:
                  >
                  > My experience resonates with D's comment here.
                  > I have increased team productivity (value delivery) many folds... to
                  see
                  > this success vanish when I left a company.
                  > While this guarantees good paychecks, experienced leaders move out of
                  > companies for better opportunities, retirement, illness etc.
                  >
                  > Most agile implementations focus on the practices which lead to
                  > fantastic short term benefits but don't hold over time.
                  > I guess focusing on organization maturity is about creating an
                  > enterprise wide (and deep) culture that does not count on a few leaders
                  > to keep progress going, but teaches continuous improvement all around.
                  > It is a slower, harder yet sustainable approach.
                  >
                  > In my new organization, most teams started with scrum and some XP
                  > colors, some switched to kanban, some not.
                  > While we still have a lot of work to do, we are not taking anything for
                  > granted, and allow ourselves to experiment around anything (reporting
                  > structure, iterations length / no iteration, rotating roles, backlog
                  > size, you name it).
                  > Many teams fake doing agile by copycatting scrum, not doing daily stand
                  > ups, retrospective, and running mini waterfall so someone can tell
                  their
                  > boss they are now agile.
                  > I think we're doing doing something different. Trying to really
                  create a
                  > change structure at all levels. My departure from my company will be
                  the
                  > ultimate test to the success I guess.
                  >
                  > Any suggestion to evaluate if the culture actually is what we try to
                  > create, before I actually leave?
                  >
                  >
                  > >I have not heard of one that existed
                  > >without the direct leadership of one of the personalities in our
                  > >community. At this point, it is impossible to take the "David" or
                  > >"Jeff" or "Israel" factor out of the achievement of the high maturity.
                  >
                Your message has been successfully submitted and would be delivered to recipients shortly.